It is well known that memory functioning suffers some deterioration with age, yet it is not entirely clear why this is the case. The proposed research is designed to investigate some underlying memory mechanisms which may be responsible for this age-related deficit. Specifically, the theoretical mechanisms of automatic spreading activation in the memory network, and attentional capacity in retrieving information from memory will be the focus of this research. The first series of experiments will address age-related differences in various characteristics of spreading activation (e.g., rate of buildup, decay, depth) in the memory network. This will be accomplished via a semantic priming task in which a prime item (e.g., cat) is presented followed by a target item (e.g., dog) and the latency to pronounce the target item is measured. The semantic relationship between the prime and target is varied, and response latency is used as an index of activation in the memory network. The second series of experiments will address age-related differences in the focusing of attention to specific areas of memory. This again will be accomplished in a semantic priming task in which the direction of attention is manipulated by varying the subjects' expectancies about which area of memory will be necessary for retrieval. In the third series of experiments, age-related differences in the retrieval of information from memory will be addressed by instantiating an episodic memory network in the context of the experiment. Subjects will be required to learn a list of sentences related to a topic. The number of sentences related to a topic, the integration of the memorized sentences, and the conditions under which the sentences are learned will be varied. The time to recognize the memorized sentences will be used as a measure of retrieval from an episodic memory network.